Focal position estimation system, focal position estimation method, focal position estimation program, semiconductor inspection system and biological observation system

Abstract

A focal position estimation system is a system for estimating a focal position when in focus corresponding to an estimation target image, and includes: an estimation target image acquisition unit that acquires an estimation target image; and a focal position estimation unit that outputs a feature quantity of the estimation target image from the estimation target image by using a feature quantity output model and estimates a focal position when in focus corresponding to the estimation target image from the output feature quantity, wherein the feature quantity output model is generated by machine learning from a plurality of learning images associated with focal position information related to a focal position at the time of imaging, and feature quantities of two different learning images are compared with each other according to focal position information associated with the two different learning images, and machine learning is performed based on the comparison result.

1 . A focal position estimation system for estimating a focal position when in focus corresponding to an estimation target image, comprising circuitry configured to:
 acquire an estimation target image; and   output a feature quantity of the estimation target image from the acquired estimation target image by using a feature quantity output model, to which information based on an image is input and which outputs a feature quantity of the image, and estimate a focal position when in focus corresponding to the estimation target image from the output feature quantity,   wherein the feature quantity output model is generated by machine learning from a plurality of learning images associated with focal position information related to a focal position at the time of imaging, and feature quantities of two different learning images are compared with each other according to focal position information associated with the two different learning images, and machine learning is performed based on a result of the comparison.   
     
     
         2 . The focal position estimation system according to  claim 1 ,
 wherein the circuitry estimates a focal position when in focus corresponding to the estimation target image from the feature quantity output from the feature quantity output model by using a focal position estimation model for estimating a focal position when in focus corresponding to an image related to a feature quantity, and   the focal position estimation model is generated by machine learning from in-focus position information related to a focal position when in focus corresponding to each of the learning images.   
     
     
         3 . The focal position estimation system according to  claim 1 ,
 wherein the circuitry controls a focal position when imaging an imaging target based on the estimated focal position.   
     
     
         4 . A semiconductor inspection system, comprising:
 the focal position estimation system according to  claim 1 ;   a mounting unit on which a semiconductor device is mounted as an imaging target related to the focal position estimation system; and   an inspection unit for inspecting the semiconductor device.   
     
     
         5 . A biological observation system, comprising:
 the focal position estimation system according to  claim 1 ;   a mounting unit on which a biological sample is mounted as an imaging target related to the focal position estimation system; and   an observation unit for observing the biological sample.   
     
     
         6 . A focal position estimation method for estimating a focal position when in focus corresponding to an estimation target image, comprising:
 acquiring an estimation target image; and   outputting a feature quantity of the estimation target image from the acquired estimation target image by using a feature quantity output model, to which information based on an image is input and which outputs a feature quantity of the image, and estimating a focal position when in focus corresponding to the estimation target image from the output feature quantity,   wherein the feature quantity output model is generated by machine learning from a plurality of learning images associated with focal position information related to a focal position at the time of imaging, and feature quantities of two different learning images are compared with each other according to focal position information associated with the two different learning images, and machine learning is performed based on a result of the comparison.   
     
     
         7 . The focal position estimation method according to  claim 6 ,
 wherein, a focal position when in focus corresponding to the estimation target image is estimated by using a focal position estimation model to which the feature quantity output from the feature quantity output model is input and which estimates a focal position when in focus corresponding to an image related to the feature quantity, and   the focal position estimation model is generated by machine learning from in-focus position information related to a focal position when in focus corresponding to each of the learning images.   
     
     
         8 . The focal position estimation method according to  claim 6 , further comprising:
 controlling a focal position when imaging an imaging target based on the estimated focal position.   
     
     
         9 . A non-transitory computer-readable storage medium storing a focal position estimation program causing a computer to operate as a focal position estimation system for estimating a focal position when in focus corresponding to an estimation target image, the focal position estimation program causing the computer to:
 acquire an estimation target image; and   output a feature quantity of the estimation target image from the acquired estimation target image by using a feature quantity output model, to which information based on an image is input and which outputs a feature quantity of the image, and estimate a focal position when in focus corresponding to the estimation target image from the output feature quantity,   wherein the feature quantity output model is generated by machine learning from a plurality of learning images associated with focal position information related to a focal position at the time of imaging, and feature quantities of two different learning images are compared with each other according to focal position information associated with the two different learning images, and machine learning is performed based on a result of the comparison.   
     
     
         10 . The non-transitory computer-readable storage medium according to  claim 9 ,
 wherein the focal position estimation program causes the computer to estimate a focal position when in focus corresponding to the estimation target image by using a focal position estimation model to which the feature quantity output from the feature quantity output model is input and which estimates a focal position when in focus corresponding to an image related to the feature quantity, and   the focal position estimation model is generated by machine learning from in-focus position information related to a focal position when in focus corresponding to each of the learning images.   
     
     
         11 . The non-transitory computer-readable storage medium according to  claim 9 ,
 wherein the focal position estimation program causes the computer to control a focal position when imaging an imaging target based on the estimated focal position.